Zurich Open Repository and Archive University of Zurich Main Library Strickhofstrasse 39 CH-8057 Zurich www.zora.uzh.ch

Year: 2016

Performance of copan WASP for routine urine microbiology

Quiblier, Chantal ; Jetter, Marion ; Rominski, Mark ; Mouttet, Forouhar ; Böttger, Erik C ; Keller, Peter M ; Hombach, Michael

Abstract: This study compared manual workup of urine clinical samples with fully automated WASPLab processing. As a first step two different inocula (1 and 10 ฀l) and different streaking patterns were compared using WASP and InoqulA BT™ instrumentation. A 10 ฀l inoculum produced significantly more single colonies than a 1 ฀l inoculum and automated streaking yielded significantly more single colonies as compared to manual streaking on whole plates (p<0.001). In a second step, 379 clinical urine samples were evaluated using WASP and manual workup. Average numbers of detected morphologies, recovered species, and CFU/ml of all 379 urine samples showed excellent agreement of WASPLab and manual workup. The percentage of clinical categorization of urine samples as ”positive” or ”negative” did not differ between automated and manual work-flow but within the positive samples automated processing by WASPLab resulted in the detection of more potential pathogens. In summary, the present study demonstrates that i) the streaking pattern, i.e. primarily the number of zigzags/length of streaking lines, is critical for optimizing the number of single colonies yielded from primary cultures of urine samples, ii) automated streaking by the WASP instrument was superior to manual streaking regarding the number of single colonies yielded, (for 32.2%) iii) automated streaking leads to higher numbers of detected morphologies (for 47.5%), species (for 17.4%) and pathogens (for 3.4%). The results of this study point to an improved quality of microbiological analyses and laboratory reports when using automated sample processing by WASP and WASPLab.

DOI: https://doi.org/10.1128/JCM.02577-15

Posted at the Zurich Open Repository and Archive, University of Zurich ZORA URL: https://doi.org/10.5167/uzh-120648 Journal Article Accepted Version

Originally published at: Quiblier, Chantal; Jetter, Marion; Rominski, Mark; Mouttet, Forouhar; Böttger, Erik C; Keller, Peter M; Hombach, Michael (2016). Performance of copan WASP for routine urine microbiology. Journal of Clinical Microbiology, 54(3):585-592. DOI: https://doi.org/10.1128/JCM.02577-15 JCM Accepted Manuscript Posted Online 16 December 2015 J. Clin. Microbiol. doi:10.1128/JCM.02577-15 Copyright © 2015, American Society for Microbiology. All Rights Reserved.

1 Performance of Copan WASP for Routine Urine Microbiology

2

3 Authors:

4 Chantal Quiblier,a* Marion Jetter,a* Mark Rominski,a Forouhar Mouttet,a Erik C. Böttger,a

5 Peter M. Keller,a° Michael Hombacha°+

6

7 * Both authors contributed equally as first authors.

8 ° Both authors contributed equally as last authors.

9

10 Authors‘ affiliation:

11 a Universität Zürich, Institut für Medizinische Mikrobiologie, Gloriastrasse 30/32, CH-

12 8006 Zürich, Switzerland.

13

14 + Corresponding Author:

15 Dr. Michael Hombach, Universität Zürich, Institut für Medizinische Mikrobiologie,

16 Gloriastrasse 30/32, CH-8006 Zürich, Switzerland, email: [email protected],

17 phone: +41 44 6342700.

18

19 Word Count: 3144

20

21 Keywords: Automated microbiology, urine, Copan, WASP.

22 Abstract

23 This study compared manual workup of urine clinical samples with fully automated

24 WASPLab processing. As a first step two different inocula (1 and 10 μl) and different

25 streaking patterns were compared using WASP and InoqulA BTTM instrumentation. A 10

26 μl inoculum produced significantly more single colonies than a 1 μl inoculum and

27 automated streaking yielded significantly more single colonies as compared to manual

28 streaking on whole plates (p<0.001). In a second step, 379 clinical urine samples were

29 evaluated using WASP and manual workup. Average numbers of detected

30 morphologies, recovered species, and CFU/ml of all 379 urine samples showed

31 excellent agreement of WASPLab and manual workup. The percentage of clinical

32 categorization of urine samples as “positive” or “negative” did not differ between

33 automated and manual work-flow but within the positive samples automated processing

34 by WASPLab resulted in the detection of more potential pathogens. In summary, the

35 present study demonstrates that i) the streaking pattern, i.e. primarily the number of

36 zigzags/length of streaking lines, is critical for optimizing the number of single colonies

37 yielded from primary cultures of urine samples, ii) automated streaking by the WASP

38 instrument was superior to manual streaking regarding the number of single colonies

39 yielded, (for 32.2%) iii) automated streaking leads to higher numbers of detected

40 morphologies (for 47.5%), species (for 17.4%) and pathogens (for 3.4%). The results of

41 this study point to an improved quality of microbiological analyses and laboratory

42 reports when using automated sample processing by WASP and WASPLab.

Evaluation of Copan WASP for Urine Microbiology 2

43 Introduction

44 In recent years, clinical microbiology has been faced with dramatic changes, as

45 full laboratory automation (FLA) has started to enter diagnostic laboratories. This trend

46 to automation will affect to various extents economic efficiency, standardization, and

47 time-to-result of laboratory procedures (16). To date, clinical microbiology is still

48 predominately based on manual sample processing. Compared to clinical chemistry,

49 microbiological specimens display a significantly higher degree of complexity (2). Thus,

50 for many years the general perception was that clinical microbiology would be far too

51 complex to allow for automated processing, and that robots would not be able to replace

52 human operators. However, it has been demonstrated that automated inoculation of

53 samples can be superior to manual processing and that automated reading of disk

54 diffusion agar plates significantly increases precision of results (4, 5, 9, 13). Thus,

55 automated sample processing promises an improved standardization of sample

56 processing, incubation times, and plate reading protocols.

57 FLA systems have been developed by several companies. Currently, there are

58 two FLA solutions available, i.e. BD Kiestra total laboratory automation (TLATM; BD

59 Kiestra B.V., Drachten, Netherlands) and WASPLab (Copan Italia S.p.A., Brescia, Italy).

60 These systems use robotic systems for handling specimen containers and primary

61 culture inoculation specifically the Kiestra InoqulA and the Copan WASP. The

62 inoculated media are moved to automated incubators by conveyor belts with integrated

63 camera systems to capture plate images at given time points which is henceforth

64 referred to in this article as WASPLab. Currently, digital plate reading (DPR) is still

65 depending on highly skilled technologists, who can read plates “virtually” without

Evaluation of Copan WASP for Urine Microbiology 3

66 physical interaction (15). First studies on FLA solutions suggest that productivity

67 indicators can be improved and diagnostic processes can be accelerated (3, 6, 7, 14).

68 In the present study, we compared workflows for urine sample processing using

69 the fully automated WASP and WASPLab systems and manual standard procedures.

70 Inoculation procedures critically influence further procedures such as specimen

71 identification (ID) and antimicrobial susceptibility testing (AST), mainly by the ability to

72 generate single colonies suitable for further processing. Therefore, both streaking

73 pattern and inoculation volume should be well evaluated prior to compare total manual

74 and automated workflows. Different guidelines exist for urine specimen inoculation, e.g.

75 U.S. and Canadian laboratories routinely plate 1 μl of urine specimens, whereas

76 European guidelines suggest 10 μl (5, 18). Therefore, this study was subdivided in two

77 parts: In part one, the ability to generate single colonies was compared for i) manual

78 and automated workflows and ii) for different automated streaking patterns using pure

79 and mixed cultures and different inocula (1 μl and 10 μl) in order to determine the

80 optimal streaking pattern. Part two of the present study compared total manual and

81 automated workflows regarding i) the number of detected CFU, ii) the number of

82 detected morphologies, iii) the number of recovered species, and iv) the number of

83 follow-up testing after initial plate reading, i.e. the number of identifications and

84 susceptibility tests, in 379 clinical urine samples.

85

86 Materials and Methods

87 Bacterial strains, clinical samples and growth conditions. Urine samples and

88 bacterial strains isolated from patient specimens used in this study were collected from

Evaluation of Copan WASP for Urine Microbiology 4

89 January until April 2015 in the clinical laboratory of the Institute of Medical Microbiology,

90 University of Zurich. If not stated otherwise bacterial cultures were incubated at 35 °C

91 ±2 °C and 7.5% CO2 for 16h-20h.

92 Quantitative analysis of streaking pattern/single colony count. Manual, semi and

93 fully automated quantitative streaking patterns were evaluated by counting the number

94 of single colonies produced on whole plates. Manual streaking was always performed

95 by the same investigator with a 1 µl and a 10 µl calibrated plastic inoculation loop

96 (Copan Italia S.p.A., Brescia, Italy) resulting in 9-10 zigzag lines (Figure 3A). For the

97 semi-automated streaking, a “ball-based” benchtop InoqulA BTTM (BD Kiestra,

98 Drachten, Netherlands) was used with a 1 µl and 10 µl inoculum and the zigzag liquid

99 streaking pattern resulting in 20 zigzag lines (Figure 3A). In this study, 1 µl and 10 µl

100 inocula were manual pipetted for InoqulA BT. Furthermore, 1 µl was taken in

101 consideration as per term of comparison but it has to be noted that the BD Kiestra

102 solutions cannot handle 1 µl. The fully automated streaking was performed by the “loop-

103 based” WASP (Copan Italia S.p.A.) with two different streaking patterns; the single

104 streak type 2 (SST2, 11 zigzag lines) and the single streak type 6 (SST6, 27 zigzag

105 lines, Figure 3A). Bacterial suspensions of McFarland 0.5, corresponding to ~1.5 x 108

106 colony forming units per ml (CFU/ml), were serial diluted in sodium borate/formate

107 containing BD Vacutainer (BD, Franklin Lakes, NJ, USA) and plated onto Columbia 5 %

108 sheep blood agar (COS, bioMérieux SA, Marcy l’Etoile, France) for monomicrobial

109 cultures or chromogenic UriSelectTM 4 agar (URI4, Bio-Rad Laboratories, Hercules, CA,

110 USA) for polymicrobial cultures. Pure Escherichia coli and enterococci cultures were

111 diluted to 104-106 CFU/ml and mixed cultures were prepared withE. coli ranging from

Evaluation of Copan WASP for Urine Microbiology 5

112 104 to 106 CFU/ml together with 106 enterococci and 106 coagulase negative

113 staphylococci (CoNS). The single colony counting was done manually by eye by the

114 same investigator.

115 Head to head comparison of urine samples. In total, 379 clinical urine samples were

116 processed in parallel by the fully automated WASPLab (Copan Italia S.p.A.) workflow

117 (inoculation, incubation, image acquisition, image analysis) and the manual routine

118 procedures of our clinical laboratory. 1 µl sample inoculum used in the manual routine

119 workflow was compared to 1 µl (121 samples) and 10 µl loop (379 samples) of the

120 WASP. Urine cultures were plated onto COS, URI4 and Columbia colistin-nalidixic acid

121 (CNA) agar 5 % sheep blood (bioMérieux SA).

122 Following manual inoculation agar plates were analyzed after one and two days of

123 incubation and morphologies, CFU count, and recovered species were recorded.

124 Material was reported negative when there was no growth or growth of a non-relevant

125 pathogen after two days of incubation. Materials were reported positive, if there was a

126 relevant pathogen in a relevant quantity, and if treatment was indicated. The sample

127 evaluation algorithm used resembles the diagnostic approach as suggested by the

128 Cumitech guidelines (12). Classification of the relevant pathogens was done as reported

129 elsewhere (10).

130 For WASP inoculation, 1 µl and 10 µl volumes of the urine sample were streaked with

131 the SST6 streaking pattern. Images were taken using WASPLab imaging after 16 and

132 38 h incubation at 35 °C ±2 °C and 7.5% CO2 and analyzed by the same laboratory

133 personnel. After 16 h photos were read, morphologies (i.e. visually different colony

134 types), and CFU counts were noted. Plates were re-read after 38 h incubation. Colonies

Evaluation of Copan WASP for Urine Microbiology 6

135 with additional morphology were identified. Samples were categorized as negative and

136 positive according to the same criteria as for the manual reading procedure. Besides the

137 number of different morphologies, CFU counts and recovered species (i.e. taxonomic

138 entities as reported by MALDI-TOF-MS identification), the number of follow-up tests,

139 such as susceptibility tests and matrix-assisted laser desorption ionization-time of flight

140 mass spectrometry (MALDI-TOF MS, Bruker Daltonics, Bremen, Germany)

141 identifications, were recorded for comparison.

142 Plate reading of manually examined specimens was performed by different laboratory

143 personnel by visual inspection directly; plates of WASP-prepared samples were read by

144 a single technician with the WASPLab, who also did the head-to-head comparison of

145 the final dataset.

146 Technical settings of WASPLab. Automated images of COS and CNA plates were

147 recorded using three different light settings (Top-light, Dark-field, and a combination of

148 Top and Back-light). URI4 images were recorded with Top-light settings only. The

149 WASPLab installation used WASPCore version 3.1.0. The settings were conserved

150 throughout the whole study duration.

151 Statistical analysis. Statistical analysis was performed using the IBM SPSS statistic

152 software, version 20 (SPSS Inc., Chicago, IL, USA). The normality of the data was

153 assessed with the Kolmogorov-Smirnov and Shapiro-Wilk test, followed by the paired t-

154 test and the non-parametric Wilcoxon signed rank test for the comparison of loop size

155 and streaking pattern for pure cultures. A cutoff value of 0.05 was considered as

156 statistically significant. For the mixed cultures, overall differences were tested using the

157 Kruskal-Wallis and the one way ANOVA with Bonferroni post hoc test to control for

Evaluation of Copan WASP for Urine Microbiology 7

158 alpha inflation. Non-parametric Follow-up tests were conducted using the Mann-

159 Whitney U test for pairwise comparison with Bonferroni correction applied for multiple

160 comparisons with an alpha level of 0.05 divided by possible number of comparisons i.e.

161 0.0083 (=0.05/6) for streaking pattern and 0.0167 (=0.05/3) for concentration as

162 independent variables. To compare the proportion between manual and WASP

163 processing of the head to head comparison the Wilcoxon signed rank test and a

164 crosstab chi-square test of independence were performed. Proportions do differ

165 significantly from each other at the level of 0.05.

Evaluation of Copan WASP for Urine Microbiology 8

166 Results

167 As a first step two different inocula (1 µl and 10 µl) and different streaking

168 patterns (one manual, two fully automated WASP/”loop-based”, and one semi-

169 automated InoqulA/”ball-based”) were evaluated with pure cultures of E. coli and

170 enterococci using CFU counts of 4/ml,10 105/ml, and 106/ml, with the view to identify the

171 loop-streaking pattern combination which yields the highest number of single colonies

172 isolated (Figure 2, iguresF S1 and S2).

173 We also studied, mixed inocula containing6 10CFU/ml enterococci, 106 CFU/ml

174 CoNS, andE. coli with CFU/ml ranging from 104 CFU/ml to 106 CFU/ml with the view to

175 mimic mixed clinical samples of a leading pathogen and residual flora.

176 Evaluation of the optimal inoculation volume

177 The 10 µl inoculum produced significantly more single colonies than the 1 µl

178 inoculum on whole plates for both E. coli and enterococci when using CFU counts

179 between 104/ml and 105/ml. This finding was independent from the streaking pattern

180 (p<0.001, Figure 2, B and C). In contrast, no significantly higher recovery of single

181 colonies was observed for E. coli CFUs of 106/ml when using a 10 µl versus 1 µl

182 inoculum and when comparing manual streaking and the InoqulA (p=0.820, and

183 p=0.713, respectively) (Figure 2, B and C). However, WASP inoculation retained a

184 significantly higher recovery of single colonies when applying the 10 µl inoculum

185 (p<0.001, Figure 2, B and C) and manual streaking.

186 Analysis of mixed cultures of E. coli, enterococci, and CoNS basically resembled

187 the results of pure cultures (Figure 3): The 10 µl inoculum resulted in the recovery of

Evaluation of Copan WASP for Urine Microbiology 9

188 significantly more single colonies than the 1 µl inoculum for manual streaking, WASP

189 SST6 and the InoqulA (p<0.001).

190 Evaluation of the optimal streaking pattern

191 For pure cultures of E. coli and enterococci the WASP streaking pattern SST2

192 (11 zigzag) produced less single colonies than manual streaking for 6 of the 12

193 inoculum-CFU combinations, thereof 5 significantly (p<0.001, Figure 2, B and C). In

194 contrast, WASP streaking pattern SST6 (27 zigzag) and InoqulA (20 zigzag) produced

195 significantly more single colonies than manual streaking for the majority of inoculum-

196 CFU combinations (p<0.001, Figure 2, B and C). For 8 out of 12 inoculum/CFU

197 combinations the WASP SST6 streaking pattern produced significantly more single

198 colonies as compared to InoqulA, WASP SST2 or manual streaking (p<0.001, Figure 2,

199 B and C).

200 For mixed cultures of E. coli, enterococci, and CoNS both WASP SST6 and

201 InoqulA yielded significantly more single colonies as comparedto manual streaking

202 (p<0.001). No significant difference in the recovery of single colonies was found for

203 mixed inocula when comparing WASP SST6 and InoqulA (p=0.071) for the 10 µl

204 inoculum, whereas for the 1 µl inoculum InoqulA BT produced significantly more

205 colonies (p=0.001).

206 Head to head comparison of clinical samples

207 To study the performance of the WASP and WASPLab in the routine

208 microbiological work up we evaluated fully automated WASP processing (streaking,

209 incubation and reading) with manual processing for 379 clinical urine samples. Based

210 on the evaluation of optimal inoculation and streaking pattern (see above), a 10 µl Evaluation of Copan WASP for Urine Microbiology 10

211 inoculum combined with SST6 (27 zigzag) streaking pattern was chosen for this

212 evaluation. We found an agreement between automated WASP processing and manual

213 workup for detected morphologies, recovered species, and CFU/ml, in 177 (46.7%), 287

214 (75.7%), and 239 (63.1%), of all 379 samples. However, WASP yielded significantly

215 higher numbers of morphologies and species recovered as compared to manual

216 processing in 180 (47.5%), and 66 (17.4%) of the 379 samples (p<0.001, Table 1,

217 Figure 4, A and B). For 122 (32.2%) of the 379 samples, WASP workup (10 µl

218 inoculum) indicated a higher CFU count/ml than manual handling (mean increase of

219 1 log10).

220 To exclude a bias due to the higher inoculum used in WASP processing, as

221 compared to manual processing, a subset of 121 samples were WASP-inoculated using

222 a 1 µl loop. No significant difference in CFU count/ml was detected between automated

223 and manual workup, when inoculum was equally set to 1 µl (p=0.176, Table 2, Figure 4

224 C). However, the percentage of samples which displayed higher numbers of

225 morphologies and higher numbers of recovered species was still significantly higher

226 when using automated sample processing, as compared to manual workup (42/121, i.e.

227 34.7%, and 18/121, i.e. 14.9% of samples, respectively, p<0.001, Table 2, Figure 4 A

228 and B).

229 Automated processing lead to the detection of more potential pathogens with

230 both the 10 µl and the 1 µl inoculum (3.4% and 9.1% of all 379 and 121 urine samples

231 processed with the corresponding inoculum, respectively, Table 2).

232 The percentage of urine samples categorized in clinical reports as “positive”, i.e.

233 treatment indicated, or “negative”, i.e. no treatment indicated, did differ but not

Evaluation of Copan WASP for Urine Microbiology 11

234 statistically significantly between manual or automatic handling. Overall, 37.2% and

235 40.4% of 379 samples were categorized as “positive” for manual and automated workup

236 respectively, while 62.8 and 59.6% were categorized as “negative” (Table 1). This ratio

237 was not affected by using different loop sizes for inoculation (compare to Table 2).

238 The significantly higher number of morphologies detected by WASPLab

239 processing lead to significantly more MALDI-TOF based identifications. However, the

240 number of subsequent antibiotic susceptibility tests performed was found not

241 significantly different between automated and manual workup, irrespective of whether

242 inoculation was done using a 10 µl or 1 µl loop (Tables 1 and 2).

Evaluation of Copan WASP for Urine Microbiology 12

243 Discussion

244 Evaluation of the optimal inoculum

245 Traditionally, North American clinical laboratories use a 1 µl inoculum for urine

246 specimens, while European laboratories inoculate 10 µl as the primary inoculum

247 according to various guidelines (8, 11, 12, 17, 18). General arguments between the two

248 inoculation volumes are a potentially decreased sensitivity using a 1 µl inoculum and the

249 missing of pathogens due to a potential lack of single colonies using a 10 µl inoculum

250 for high CFU counts. We aimed at determining the optimal inoculum for the majority of

251 CFU/pathogen combinations that are most frequently encountered in the clinical

252 laboratory. E. coli was chosen as the most prevalent uropathogen (1). Enterococciand

253 CoNS were selected to represent the most frequently encountered Gram-positive

254 species in urine specimens (10). On whole plates the 10 μl inoculum was demonstrated

255 to be superior to the 1 μl inoculum in order to maximize the equencyfr of single colony

256 recovery, particularly for CFUs of ≤106/ml. As positivity rates and the number of

257 potential pathogens did not differ between the 1 µl and the 10µl inoculum, the

258 abovementioned concerns on decreased sensitivity (1 µl inoculum) and missing of

259 pathogens (10 µl inoculum) could not be confirmed. A higher frequency of single colony

260 recovery using automated inoculation may have contributed to the increased detection

261 of potential pathogens (see also “Head to head comparison of clinical samples”).

262 However, the effect of single colony recovery and improved visual inspection by high

263 resolution imaging on sample categorization cannot be distinguished in this study.

Evaluation of Copan WASP for Urine Microbiology 13

264 Evaluation of the optimal streaking pattern

265 For pure cultures, WASP SST6 streaking was superior to both WASP SST2 and

266 InoqulA streaking with respect to maximize the recovery of single colonies,

267 independently of the inoculum volume. For mixed cultures, single colony counts

268 generated by WASP SST6 and InoqulA streaking did not differ significantly using a 10 µl

269 inoculum (p=0.071). However, the number of samples investigated was limited, and a

270 larger follow up study may reveal statistically significant results given the borderline p-

271 value. The bench-top InoqulA in combination with manual pipetting produced more

272 colonies as compared to the SST6 for the 1 µl inoculum (p=0.001). However, the fully

273 automated InoqulA cannot handle inoculation volumes lower than 10 µl. In part, these

274 findings are in contrast to a recent study that reported the InoqulA to be superior to

275 WASP streaking with respect to the recovery of single colonies (4, 5, 13). In our study,

276 this effect was shown mainly for exceptionally high CFU counts of 107 to 108 CFU/ml.

277 However, as the vast majority of clinical specimens (228/306 = 74.5% in this study)

278 contain up to 105 CFU/ml. Here, for pure cultures WASP streaking was even superior to

279 the InoqulA method, if SST6 streaking with 27 zigzags was applied. WASP SST2

280 streaking using 11 zigzags was significantly inferior in generating single colonies. Thus,

281 the critical parameter was the streaking pattern selected, i.e. number of zigzags, rather

282 than the instrument used for automated inoculation. Differences in streaking patterns

283 may be the reason for discrepancies between the results of this study and the results of

284 the studies of Croxatto, Froment, and Mischnik as discussed above (4, 5, 13).

285 Both WASP SST6 and InoqulA were superior to manual streaking in maximizing

286 single colony recovery. These findings are in agreement with those of other authors (4).

Evaluation of Copan WASP for Urine Microbiology 14

287 The same study found a significant improvement of automation over manual processing

288 mainly for high CFU counts (>106/ml), the present study demonstrates a significant

289 increase in recovery of single colonies using automated inoculation for CFU counts

290 down to104 CFU/ml (4).

291 Head to head comparison of clinical samples

292 WASP processing of clinical specimens yielded significantly more morphologies

293 and species recovered as compared to manual processing, independently from the

294 number of CFU counts present in the sample. This finding parallels those of a previous

295 study that analyzed urine specimens processed with the BD-Kiestra system versus

296 manual handling (5, 18). Indeed, next to a higher number of morphologies and species

297 isolated, automated processing by WASP lead to detection of more potential pathogens.

298 Other authors evaluating the BD-Kiestra system with urine specimens did not find more

299 pathogens; however, plate reading was not done by camera but by the unaided eye

300 (18). The high resolution of the WASPLab imaging system may essentially contribute to

301 the higher number of morphologies detected, resulting in a higher number of identified

302 species and pathogens.

303 The percentage of categorization of urine samples in clinical reports as “positive”,

304 i.e. treatment indicated, or “negative”, i.e. no treatment indicated, did differ but not

305 statistically significant between automated or manual handling. These results are in

306 agreement with data of other studies evaluating the BD-Kiestra system (18).

307 The results of the head to head comparison of clinical samples demonstrated

308 that automation in urine microbiology is feasible. However, despite a higher number of

309 potential pathogens within “positive” samples in the automated workup, the overall

Evaluation of Copan WASP for Urine Microbiology 15

310 categorization of samples as positive or negative was not significantly different between

311 automated and manual processing.

312 A limitation of this study is that a reliable evaluation of the time to result was not

313 possible, because administrative and organizational lab procedures (for instance no

314 24h/7d service) must be adapted during the process of automated workflow

315 implementation as the influence of these factors on time to result is significant. As the

316 system was primarily evaluated technically, and WASPLab was not part of the

317 diagnostic routine workflow yet, reliable conclusions regarding time to result could not

318 be drawn from this study and should be subject to a follow up work. There are three

319 main conclusions that can be drawn from the present study: i) The streaking pattern, in

320 particular the number of zigzags/length of streaking lines is of critical importance for

321 increasing the number of single colonies recovered from primary cultures; ii) Automated

322 streaking by WASP can significantly increase the number of single colonies recovered,

323 iii) Automated WASPLab processing results in a higher number of detected

324 morphologies, species, and pathogens. In summary, our findings demonstrate that

325 automated sample processing by WASPLab is going to improve the quality of urine

326 microbiological analysis.

Evaluation of Copan WASP for Urine Microbiology 16

327 References

328 1. Barber, A. E., J. P. Norton, A. M. Spivak, and M. A. Mulvey. 2013. Urinary

329 tract infections: current and emerging management strategies. Clinical infectious

330 diseases 57:719-724.

331 2. Bourbeau, P. P., and N. A. Ledeboer. 2013. Automation in clinical microbiology.

332 Journal of Clinical Microbiology 51:1658-1665.

333 3. Bourbeau, P. P., and B. L. Swartz. 2009. First evaluation of the WASP, a new

334 automated microbiology plating instrument. Journal of Clinical Microbiology

335 47:1101-1106.

336 4. Croxatto, A., K. Dijkstra, G. Prod'hom, and G. Greub. 2015. Comparison of

337 the InoqulA and the WASP automated systems with manual inoculation. Journal

338 of Clinical Microbiology:JCM. 03076-03014.

339 5. Froment, P., H. Marchandin, P. V. Perre, and B. Lamy. 2014. Automated

340 versus manual sample inoculations in routine clinical microbiology: a

341 performance evaluation of the fully automated InoqulA instrument. Journal of

342 Clinical Microbiology 52:796-802.

343 6. Greub, G., and G. Prod'hom. 2011. Automation in clinical bacteriology: what

344 system to choose? Clinical microbiology and infection : the official publication of

345 the European Society of Clinical Microbiology and Infectious Diseases 17:655-

346 660.

347 7. Grohs, P., B. Tillecovidin, A. Caumont-Prim, E. Carbonnelle, N. Day, I.

348 Podglajen, and L. Gutmann. 2013. Comparison of five media for detection of

Evaluation of Copan WASP for Urine Microbiology 17

349 extended-spectrum Beta-lactamase by use of the wasp instrument for automated

350 specimen processing. Journal of Clinical Microbiology 51:2713-2716.

351 8. Gupta, K., T. M. Hooton, K. G. Naber, B. Wullt, R. Colgan, L. G. Miller, G. J.

352 Moran, L. E. Nicolle, R. Raz, A. J. Schaeffer, and D. E. Soper. 2011.

353 International clinical practice guidelines for the treatment of acute uncomplicated

354 cystitis and pyelonephritis in women: A 2010 update by the Infectious Diseases

355 Society of America and the European Society for Microbiology and Infectious

356 Diseases. Clinical infectious diseases : an official publication of the Infectious

357 Diseases Society of America 52:e103-120.

358 9. Hombach, M., R. Zbinden, and E. C. Bottger. 2013. Standardisation of disk

359 diffusion results for antibiotic susceptibility testing using the sirscan automated

360 zone reader. BMC microbiology 13:225.

361 10. Hooton, T. M. 2012. Clinical practice. Uncomplicated urinary tract infection. The

362 New England journal of medicine 366:1028-1037.

363 11. Jorgensen, J. H., M. A. Pfaller, K. C. Carroll, G. Funke, M. L. Landry, S. S.

364 Richter, and D. W. Warnock. 2015. Manual of Clinical Microbiology, 11th

365 Edition. American Society of Microbiology.

366 12. McCarter, Y., E. Burd, G. Hall, and M. Zervos. 2009. Cumitech 2C: laboratory

367 diagnosis of urinary tract infections. American Society for Microbiology,

368 Washington, DC.

369 13. Mischnik, A., M. Mieth, C. J. Busch, S. Hofer, and S. Zimmermann. 2012.

370 First evaluation of automated specimen inoculation for wound swab samples by

371 use of the Previ Isola system compared to manual inoculation in a routine

Evaluation of Copan WASP for Urine Microbiology 18

372 laboratory: finding a cost-effective and accurate approach. Journal of Clinical

373 Microbiology 50:2732-2736.

374 14. Mutters, N. T., C. J. Hodiamont, M. D. de Jong, H. P. Overmeijer, M. van den

375 Boogaard, and C. E. Visser. 2014. Performance of Kiestra total laboratory

376 automation combined with MS in clinical microbiology practice. Annals of

377 laboratory medicine 34:111-117.

378 15. Rhoads, D., S. Novak, and L. Pantanowitz. 2015. A review of the current state

379 of digital plate reading of cultures in clinical microbiology, vol. 6.

380 16. Sautter, R. L., and R. B. Thomson, Jr. 2015. Consolidated clinical microbiology

381 laboratories. Journal of Clinical Microbiology 53:1467-1472.

382 17. Sobel, J. D., and D. Kaye. 2015. 74 - Urinary Tract Infections, p. 886-913.e883.

383 In J. E. B. D. J. Blaser (ed.), Mandell, Douglas, and Bennett's Principles and

384 Practice of Infectious Diseases (Eighth Edition). Content Repository Only!,

385 Philadelphia.

386 18. Strauss, S., and P. P. Bourbeau. 2015. Impact of introduction of the BD Kiestra

387 InoqulA on urine culture results in a hospital clinical microbiology laboratory.

388 Journal of Clinical Microbiology 53:1736-1740.

389

390

391

Evaluation of Copan WASP for Urine Microbiology 19

392 Acknowledgements

393 This study was partially supported by the University of Zurich.

394 Prof. Erik C. Böttger is a consultant of Copan Italia, S.p.A.

395

Evaluation of Copan WASP for Urine Microbiology 20

396 Figures

397 Figure 1. Study design and evaluation steps of part 1 the evaluation of the optimal

398 inoculation volume and streaking pattern (A) and part 2 the head to head comparison

399 (B).

400

Evaluation of Copan WASP for Urine Microbiology 21

401 Figure 2. Streaking pattern details and resulting numbers of single colony counts. (A)

402 One manual (routine) and three automated quantitative streaking patterns with varying

403 numbers of zigzag (zz) lines shown in brackets; WASP single streak type 2 (SST2);

404 WASP single streak type 6 (SST6) and InoqulA. Comparison of single colonies obtained

405 by manual and automated streaking. 1 µl and 10 µl E. coli (B) and enterococci (C)

406 bacterial suspensions were used with CUFs ranging from 4 10to 106. The mean values

407 of 30 clinical strains per species are shown with their standard deviation.

408

409

410

411

Evaluation of Copan WASP for Urine Microbiology 22

412 Figure 3 Cumulated numbers of single colonies recovered from mixed cultures

413 containing variable CFU counts of E. coli (104 to 106 per ml as indicated in the Figure)

414 PLUS fixed CFU counts of 106 per ml of both enterococci and coagulase negative

415 staphylococci. Mean values of en t mixed cultures prepared with individual clinical

416 strains are shown with theaccording standard deviation.

417

Evaluation of Copan WASP for Urine Microbiology 23

418 Figure 4 Correlation of morphologies (A), recovered species (B), and CFU resulting (C)

419 from the head to head comparison of 121 and 379 clinical samples, respectively. Data

420 indicate a shift to higher numbers for all three parameters for automated as compared to

421 manual inoculation.

422

A Morphology Correlation

WASP WASP 1µl 10µl 01234≥5 01234≥5 Manual Manual 0 37 1 1 0 73 29 10 2

1 112111 1 30 30 8 1

2 5424 2 11818136

3 3963 3 1 5 24 29 13

4 159 4 61921

≥5 33 ≥5 2713

B Recovered species correlation

WASP WASP 1µl 10µl 01234≥501234 ≥5 Manual Manual

0 55 8 4 0 161 35 5

1 1336 1 891241

2 56 2 4930 1

3 111 3 135

44

≥5 ≥5 1

C CFU correlation

WASP WASP 1µl 10µl n. g. < 104 104 105 106 ≥106 n. g. < 104 104 105 106 ≥106 Manual Manual

n. g. 37 2 n. g. 73 40 1

< 104 1178 1 < 104 74 18 6 1

104 4142 104 11 30 19 7

105 1124 105 2 1 22 30

106 315 106 440

6 6 423 ≥10 ≥10

Evaluation of Copan WASP for Urine Microbiology 24

424 Table 1. Results of the head to head comparison of clinical samples. Manual inoculation

425 versus WASP inoculation (10 µl loop). Significant differences are indicated in bold (non-

426 parametric Wilcoxon signed rank test (p<0.05) and crosstab chi-square test of

427 independence (p<0.05).

428 429 N Samples with indicated parameter (N, % of all samples)

Manual > WASP > Parameter Manual WASP Identical WASP Manual p

Total number of samples 379

180 Morphologies 177 (46.7%) 22 (5.8%) (47.5%) 0.000

Colony forming units 122 (CFU) /ml 239 (63.1%) 14 (4.8%) (32.2%) 0.000

Recovered species 287 (75.7%) 26 (6.9%) 66 (17.4%) 0.000

Total No. of MALDI- TOF identifications 253 313 0.000

Total No. of susceptibility tests 149 163 0.337

Report 199 (52.5%) 180 (47.5%)

Positive result 141 (37.2%) 153 (40.4%)

Negative result 238 (62.8%) 226 (59.6%)

Possible Pathogens1) 159 (42%) 172 (45.4%)

Contaminated 10 (2.6%) 10 (2.6%) 430 431 1) Possible Pathogens: Gram negative rods, Staphylococcus aureus, Staphylococcus lugdunensis, 432 Staphylococcus saprophyticus, Streptococcus group B, C or G, enterococci, 433 yeasts, Corynebacterium glucuronolyticum, Corynebacterium urealyticum

Evaluation of Copan WASP for Urine Microbiology 25

434 Table 2. Results of the head to head comparison of clinical samples. Manual inoculation 435 versus WASP inoculation (1 µl loop). Significant differences are indicated in bold 436 (non-parametric Wilcoxon signed rank test (p<0.05) and crosstab chi-square test of 437 independence (p<0.05). 438 N Samples with indicated parameter (N, % of all samples)

Manual > WASP > Parameter Manual WASP Identical WASP Manual p

Total number of samples 121

Morphologies 71 (58.7%) 8 (6.6%) 42 (34.7%) 0.000

Colony forming units (CFU) /ml 95 (78.5%) 9 (7.4%) 17 (14.1%) 0.176

Recovered species 95 (78.5%) 8 (6.6%) 18 (14.9%) 0.043

Total No. of MALDI- TOF identifications 76 100 0.001

Total No. of susceptibility tests 47 52 0.601

Report 79 (65.3%) 42 (34.7%)

Positive result 47 (39.7%) 51(42.1%)

Negative result 73 (60.3%) 70 (57.9%)

Possible 1) pathogens 48 (39.7%) 59 (48.8%)

Contaminated 6 (5%) 2 (1.7%)

439 440 441 1) Possible Pathogens: Gram negative rods, Staphylococcus aureus, Staphylococcus lugdunensis, 442 Staphylococcus saprophyticus, Streptococcus group B, C or G, enterococci, 443 yeasts, Corynebacterium glucuronolyticum, Corynebacterium urealyticum 444 445

446

Evaluation of Copan WASP for Urine Microbiology 26